Staphylococcus aureus is among the leading causes of both post traumatic and post operative endophthalmitis. Despite aggressive therapy in the form of intraocular, antibiotics, anti-inflammatory agents, and vitrectomy S. aureus endophthalmitis is frequently associated with loss of vision (62% of reported cases). Even though the pathogenicity of S. aureus in extraocular infections has been the subject of extensive research, the pathogenic factors elaborated by this organism intraocularly have not been determined. An awareness of the virulence factors contributing to severity would support the development of new, information-based, therapeutic strategies that could augment existing treatments with a view to decreasing the frequency of visual loss. The PI has proposed and initiated a new and independent study to identify factors that are responsible for the severity of S. aureus endophthalmitis. Generally, factors that contribute to tissue damage during and after infections include secreted toxins and bacterial cell surface components that induce a host inflammatory response. The large number of potential virulence factors expressed by S. aureus presents a significant challenge in identifying those factors contributing to severity of endophthalmitis. As a first step toward identifying factors contributing to severity of S. aureus endophthalmitis, the virulence traits associated with severe disease in the rabbit model will be identified. This will be done by comparing the pathogenesis of infections caused by toxin-producing wild type strains of S. aureus and attenuated strains (agr-, sar-, agr-/sar-) defective in the production of clusters of extracellular products. In addition, mutant strains defective in the production of individual toxins will be tested for their relative avirulence when compared to isogenic parental strains. The influence of non-secreted surface components of S. aureus will be analyzed by injecting heat treated, non-metabolizing wild type S. aureus into rabbit eyes and evaluating the pathogenic effects. Having identified those virulence traits that contribute to severity in the rabbit model, approximately 100 S. aureus endophthalmitis clinical isolates will be analyzed for the production of the same toxins. Statistical analyses will be used to identify the toxins for which there is a strong correlation with the documented severity of infection caused by each isolate. Once identified, these toxins will become the focus of subsequent studies that will test the validity of the rabbit model as an accurate reflection of human disease. Since virulence traits will have been identified that are associated with severe endophthalmitis in both humans and rabbits, the model will serve as an ideal tool for testing the effectiveness of new therapeutic strategies, which is the long term goal of this study.